Remove spurious peaks of Hilbert transform for LP residual

[ncomputing]

Homework Statement

I want to eliminate spurious peaks of Hilbert transform for finding Glottal closure in LP residual. I have 4 step :

Homework Equations

1-down-sample.

2-Hilbert Transform.

3-Identify Peaks in Hilbert Transform.

4-consider this hypothesis that time gap between two successive glottal closure instance is not likely to vary in the adjacent pitch period.

I have wrote following MATLAB code:

The Attempt at a Solution

e_downsample = downsample(e,2);
DFT=fft(e);
hilbert_e=ifft(complex(imag(DFT(1:(length(DFT)/2))),real(DFT(1:(length(DFT)/2)))));
h_e=sqrt(e_downsample(:).^2-hilbert_e(:).^2);
[pks,locs] = findpeaks(abs(h_e));
figure,
plot(abs(h_e));

 

[Valkarie]

hold on;plot(locs,pks,'ro');for i=1:length(locs)-1 gap_period(i)= locs(i+1)-locs(i);endmean_gap_period=mean(gap_period);[pks,locs] = findpeaks(abs(h_e));for i=1:length(locs)-1 if (abs(locs(i+1)-locs(i))>(mean_gap_period*1.5)) locs(i+1)=0; endendlocs=locs(locs ~= 0); figure,plot(abs(h_e));hold on;plot(locs,pks(locs),'ro');